Distilling knowledge from Neural Networks to build smaller and faster models

Distilling knowledge from Neural Networks to build smaller and faster models https://blog.floydhub.com/knowledge-distillation/

Fri 15 Nov 2019 08:20:21 PM CET

This article discusses GPT-2 and BERT models, as well using knowledge distillation to create highly accurate models with fewer parameters than their teachers

Links per page

Display